Shortly after midnight on September 15, an Italian fighter flying NATO’s Baltic air-policing mission shot down a drone near the village of Pratkunai, not far from Kaunas, Lithuania’s second city. The drone had entered from Belarus. Lithuania’s defense minister, Robertas Kaunas, said it carried an explosive and that only the wreckage would show whether it was Russian or “another country’s drone.” His Italian counterpart, Guido Crosetto, was less cautious, saying the threat “most probably came from Russia.” Stars and Stripes reported that a typical two-fighter NATO interception runs close to $100,000 before a missile is fired.
The shootdown came a year and three days after NATO announced Eastern Sentry, and about a year after the European Commission president proposed a “drone wall” so that fighters would not have to do this job. This article sets out what that proposal has turned into by October 2026. In brief: there is no single wall and no single commander. There are national border systems at different stages of completion, a NATO air mission that still does most of the shooting with crewed jets, and two European Union project labels that became eligible for money two weeks ago. The threat has also changed shape since the idea was drawn up, which is the part of the story that gets the least attention.
From a wall to two project labels
The starting point is not in dispute. On the night of September 9, 2025, roughly 20 Russian drones crossed into Poland, and allied aircraft shot down four of them, according to the Kyiv Independent’s reconstruction. Poland asked for consultations under Article 4 of the North Atlantic Treaty, and NATO launched Eastern Sentry within days. Ursula von der Leyen then called for a drone wall, a concept Reuters traced to a Baltic border-security scheme and to a briefing Ukraine’s Mykhailo Fedorov had given her on Kyiv’s counter-drone methods.
The idea ran into resistance within weeks. Reuters reported in October 2025 that France and Germany were unwilling to hand the Commission control of major procurement, that Emmanuel Macron called the threat “more sophisticated, more complex” than a wall implied, and that German Defense Minister Boris Pistorius doubted it could be built in three or four years. Southern and western governments complained of being left out, and the Commission widened the plan into a European Drone Defence Initiative covering the whole continent. Its Defence Readiness Roadmap set a target of a fully functional drone initiative by the end of 2027 and an operational Eastern Flank Watch by the end of 2028. Breaking Defense reported that both were meant to reach initial capacity by the end of 2026.
What exists on paper today is narrower and more industrial than that language suggested. On July 3 the Commission proposed five European Defence Projects of Common Interest, and on September 28 the Council approved them. Two matter here. One is Eastern Flank Watch. The other is a drone and counter-drone project that goes by the acronym DECODER. The approval makes the projects eligible for European Defence Industry Programme funds. It does not award any.
What the money does and does not cover
The distance between the headline figures and the committed cash is large. The Commission’s stated ambition for the five projects together is about €190 billion by 2036. Defence24 gives Eastern Flank Watch a range of €60 billion to €100 billion by 2036 and the drone project €3.5 billion to €5 billion by 2033. The EU money actually set aside to start all five is €325 million, drawn from a €1.5 billion program. The larger numbers are Commission estimates of investment need and not committed funding, as EU Perspectives noted, and the Council’s announcement points only to possible later support from a planned European Competitiveness Fund.
Membership is uneven as well. EU Perspectives reported that the drone project has 26 member states plus Norway and Ukraine, while Eastern Flank Watch has 13 member states with the same two partners. France is in four of the five projects and has stayed out of Eastern Flank Watch, according to Defence24. Governance is consistent with what Paris and Berlin were reported to want in 2025: the Commission helps and tracks progress, and individual countries carry responsibility for implementation.
A separate and smaller stream comes from the interior side of the Commission. On October 1 it added €130 million for border surveillance and counter-drone equipment under its border management instrument, bringing that pot to more than €500 million. That money is tied to an action plan on drone security launched in February, which also provides for a counter-drone center of excellence. Its stated purpose is surveillance of external borders, which is a narrower task than air defense.
Whether the end-of-2026 initial capacity target will be met is not publicly known, because none of the documents reviewed for this article defines what initial capacity consists of. Radio Free Europe’s Brussels correspondent wrote in late August that the drone wall was conspicuously absent from recent EU texts and quoted an unnamed Baltic official: “Nobody has this wall yet, not even the Ukrainians.”
The threat of 2026 is not the one the wall was drawn for
The original picture was of Russian strike drones crossing a line on a map from east to west. Much of what has happened along the northern part of that line this year does not fit it. A Reuters chronology lists a Ukrainian drone hitting the chimney of Estonia’s Auvere power station on March 25 and another exploding at an oil depot in Rezekne, Latvia, on May 7. Latvia’s defense minister resigned three days after Rezekne, and the prime minister followed four days later. NATO fighters shot down stray Ukrainian drones over Estonia in May and over Latvia in June and August.
Kyiv has apologized and says the cause is Russian interference with its long-range strikes on Russian Baltic ports. Defense News reported that Lithuania had detected 36 Russian spoofing transmitters by May, up from three at the start of 2025, and Ukraine’s foreign minister said investigations proved deliberate diversion. That last point remains Ukraine’s claim. Finland has said only that Russian jamming could explain the drift, and the origin of the drone downed over Lithuania had not been announced in the reports reviewed for this article.
Further south the pattern is closer to the 2025 one. Romanian authorities recorded more than 40 incursions in the first half of the year, a Russian drone struck an apartment block in Galați on May 29, and Romania shot down a Shahed-type drone on July 24, with three more in the weeks that followed. A third category does not cross a border at all. In early August an explosive-laden drone was found at Leipzig/Halle Airport near a Ukrainian cargo aircraft, and Germany’s interior minister attributed the operation to Russia on the basis of components and explosives. Moscow denies responsibility, and the case is with federal prosecutors. As for the wave of airport sightings in late 2025, the European Council on Foreign Relations concluded in August that it remains unproven where most of those drones came from.
NATO’s supreme allied commander, Gen. Alexus Grynkewich, told a European Parliament committee on September 28 that “the Russians have been much more careful to avoid coming across the border with drones in the Baltics or into Poland” since Eastern Sentry began, according to Reuters. If he is right, the flank from Estonia to Poland now faces largely friendly drones that have gone off course, a problem of identification and of coordination with Kyiv more than of interception.
What has been built is national
Hardware is arriving through national programs that predate or bypass Brussels. Poland signed a contract on January 30 for its SAN system, worth about 15 billion zloty, covering 18 batteries, 52 firing platoons and roughly 700 vehicles, led by state group PGZ with Norway’s Kongsberg and financed through SAFE loans. ECFR puts initial capability at the end of 2026 and full operation a year later. Neither date has been demonstrated. In the meantime Poland put its FA-50 light fighters on quick-reaction alert on October 1, primarily against jet-powered Geran drones, in the words of a Polish general quoted by AeroTime.
Latvia’s new defense minister, Raivis Melnis, told Business Insider that an automated system along the whole Russian border would be finished by the end of September, built around cheap interceptor drones, with soldiers making only the final decision to engage and fighters reduced to “the secondary means.” That is a ministerial claim. No independent account of a completed, working system was found, and Latvia still asked Germany to send four Eurofighters to guard its October 3 election. At a Latvian test range in late May, Defense News watched one interceptor hit two targets in three attempts and another fail on its first try. “We face serious problems, not only in Latvia,” said Maj. Modris Kairišs, who runs the country’s autonomous systems center.
Elsewhere the record is thinner. Estonia has installed fixed detection along sections of its border, and an Estonian-led bid for EU money for a Baltic drone wall was turned down in March 2025, ECFR notes. Lithuania is buying 48 U.S.-made Merops interceptors, and Business Insider reported that at least six NATO countries have bought or plan to buy the system. ECFR found minimal progress in Bulgaria, Hungary and Slovakia. None of these national networks is known to feed a shared picture of the whole frontier, which Estonian Defense Minister Hanno Pevkur identified last November as the hard part, saying that low-altitude detection was the most difficult element.
Ukraine supplies the method and inherits the next problem
Ukraine’s role has grown from adviser to participant. It sits in four of the five EU projects. An EU-Ukraine Drone Alliance held its first board meeting on September 11 with nine companies from each side, and Commissioner Andrius Kubilius said there that first deliverables should arrive by the end of this year. Bilateral deals are moving faster. Latvia and Ukraine signed a drone agreement on June 9 and plan a joint factory in Latvia’s eastern Latgale region, and Kyiv has agreed to send specialists to Lithuania. Analysts at the Center for European Policy Analysis argued last December that a wall without Ukraine would be a barricade short of bricks.
The method being imported is already under strain at home. Russia’s jet-powered Geran-5 cruises at 450 to 600 kilometers per hour, by AeroTime’s account, and Ukrainian firms at Poland’s MSPO trade fair in September told Defense News they were redesigning interceptors with boosters and jet engines to keep pace with it. A border system bought in 2026 around today’s interceptors may therefore be chasing a target that has already moved on.
The case that a wall was never the right test
The strongest objection to judging Europe against the wall is that the wall was a slogan, and that what has emerged instead is closer to what serious critics asked for. ECFR’s Ulrike Franke, who judged the Commission’s widened proposal to be set for failure, recommends that Brussels back the national efforts in Poland and the Baltics, pay for laggards, set common standards and build a joint operating picture with NATO. Much of that is now the declared direction. She also warns that the EU’s “hotch-potch” of overlapping projects could scatter effort.
A second objection goes further and questions the economics that motivated the whole enterprise. Justin Bronk of the Royal United Services Institute told the Kyiv Independent that sending fighters after individual drones is simpler than meeting the “huge personnel and logistics requirements of fielding thousands of individually cheap short-range systems” along an entire border, and that the priority should be cheaper munitions for the jets, such as laser-guided rockets. Federico Borsari of CEPA made the related point that a missile’s price is small beside the damage a drone can do. Jennifer Kavanagh argued in War on the Rocks in 2025 that the incursions were hybrid pressure to be met with modest, layered defenses and not with grand projects. On this reading, a year of fighters, national sensor lines and a deterrent patrol is a reasonable result, and Grynkewich’s remark about Russian caution is evidence for it.
The reply is that the fighter model depends on a low volume of intruders and on political tolerance for alarms. Latvia’s government fell after one depot fire, and Finland warned 1.8 million people in the Helsinki area to stay indoors in May. Kubilius himself told the European Parliament’s security and defense subcommittee, in remarks reported on October 7, that the EU is not rearming fast or effectively enough to be ready by 2030, according to Defence24’s account.
Several decisions are still open, and they belong to different owners. Member states negotiating the next EU budget will determine whether Eastern Flank Watch gets more than seed money, and France’s absence from it suggests that argument is not settled. NATO and the Commission have not said publicly who will own a common air picture from Finland to Bulgaria or on what data standard it will run. National parliaments and defense ministries have to decide how much of the engagement sequence can be automated, which RFE/RL listed among the points still in dispute. Kyiv and its neighbors have to work out procedures for Ukrainian strike drones that leave their planned routes. The first measurable test is close: the Commission’s own initial-capacity date falls at the end of December, and it has yet to say what will be counted.
This analysis draws on the public sources linked in the text. Send corrections to info@defenseautonomyreview.com.
